Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=tent20

Download by: [Oulu University Library] Date: 01 March 2017, At: 08:47

Environmental Technology

ISSN: 0959-3330 (Print) 1479-487X (Online) Journal homepage: http://www.tandfonline.com/loi/tent20

Removal of ammonium from municipalwastewater with powdered and granulatedmetakaolin geopolymer

Tero Luukkonen, Kateřina Věžníková, Emma-Tuulia Tolonen, Hanna Runtti,Juho Yliniemi, Tao Hu, Kimmo Kemppainen & Ulla Lassi

To cite this article: Tero Luukkonen, Kateřina Věžníková, Emma-Tuulia Tolonen, Hanna Runtti,Juho Yliniemi, Tao Hu, Kimmo Kemppainen & Ulla Lassi (2017): Removal of ammonium frommunicipal wastewater with powdered and granulated metakaolin geopolymer, EnvironmentalTechnology

To link to this article: http://dx.doi.org/10.1080/09593330.2017.1301572

Accepted author version posted online: 01Mar 2017.

Submit your article to this journal

View related articles

View Crossmark data

Publisher: Taylor & Francis & Informa UK Limited, trading as Taylor & Francis

Group

Journal: Environmental Technology

DOI: 10.1080/09593330.2017.1301572

Removal of ammonium from municipal wastewater with powdered

and granulated metakaolin geopolymer

Tero Luukkonen1, Kateřina Věžníková2, Emma-Tuulia Tolonen1,3, Hanna

Runtti3, Juho Yliniemi4, Tao Hu3, Kimmo Kemppainen1, Ulla Lassi3,5

1 Kajaani University of Applied Sciences, Kuntokatu 5, FI-87101 Kajaani, Finland

2 Faculty of Chemistry, Brno University of Technology, Purkyňova 464/118, CZ-61200

Brno, Czech Republic

3 University of Oulu, Research Unit of Sustainable Chemistry, FI-90014 University of

Oulu, Finland

4 University of Oulu, Fibre and Particle Engineering Research Unit, FI-90014

University of Oulu, Finland

5 University of Jyvaskyla, Kokkola University Consortium Chydenius, Unit of Applied

Chemistry, Talonpojankatu 2B, FI-67100, Kokkola, Finlandepartment, University, City,

Country

Telephone numbers and email addresses: Tero Luukkonen +358445353695,

teroluuk@gmail.com, tero.luukkonen@kamk.fi; Kateřina Věžníková

katerina.veznikova@gmail.com; Emma-Tuulia Tolonen +358294481617 emma-

tuulia.tolonen@oulu.fi; Hanna Runtti +358294481617 hanna.runtti@oulu.fi; Juho

Yliniemi +358294482455 juho.yliniemi@oulu.fi; Tao Hu tao.hu@oulu.fi

+35847166122; Kimmo Kemppainen +358447101050 kimmo.kemppainen@kamk.fi;

Ulla Lassi +358294481592 ulla.lassi@oulu.fi

Corresponding author: Tero Luukkonen, teroluuk@gmail.com,

tero.luukkonen@kamk.fi, +358445353695, orcid.org/0000-0002-1124-775X

Acknowledgements

This work was supported by the Finnish Funding Agency for Innovation (TEKES)

under Grant 4096/31/2014 (project GeoSorbents). The authors wish to thank Mr. Kai

Tiihonen, Ms. Marjukka Hyyryläinen, and Dr. Esther Takaluoma for assisting with the

laboratory work and analyses.

Removal of ammonium from municipal wastewater with powdered

and granulated metakaolin geopolymer

Ammonium (NH4+) removal from municipal wastewater poses challenges with

the commonly used biological processes. Especially at low wastewater

temperatures, the process is frequently ineffective and difficult to control. One

alternative is to use ion-exchange. In the present study, a novel NH4+ ion-

exchanger, metakaolin geopolymer (MK-GP), was prepared, characterized, and

tested. Batch experiments with powdered MK-GP indicated that the maximum

exchange capacities were 31.79, 28.77, and 17.75 mg/g in synthetic, screened,

and pre-sedimented municipal wastewater, respectively, according to the Sips

isotherm (R2 ≥ 0.91). Kinetics followed the pseudo-second order rate equation in

all cases (kp2 = 0.04–0.24 g mg-1 min-1, R2 ≥ 0.97) and the equilibrium was

reached within 30–90 min. Granulated MK-GP proved to be suitable for a

continuous column mode use. Granules were high-strength, porous at the surface

and could be regenerated multiple times with NaCl/NaOH. A bench-scale pilot

test further confirmed the feasibility of granulated MK-GP in practical conditions

at a municipal wastewater treatment plant: consistently < 4 mg/L NH4+ could be

reached even though wastewater had low temperature (approx. 10°C). The results

indicate that powdered or granulated MK-GP might have practical potential for

removal and possible recovery of NH4+ from municipal wastewaters. The simple

and low-energy preparation method for MK-GP further increases the significance

of the results.

Keywords: Alkali activation; ammonium; geopolymer; ion exchange; municipal

wastewater

Introduction

Ammonium (NH4+) is the most important nitrogen-species contributing to the

eutrophication of water bodies in the areas where nitrogen is the nutrient in shortest

supply [1]. Consequently, nitrogen removal from municipal wastewater has become

commonly mandatory and, for example, in Europe it is specified in the urban

wastewater treatment directive [2]. The most widely used method to remove nitrogen

from wastewater is by microbial nitrification-denitrification reactions which can be

implemented by, for instance, the aerobic-anoxic active sludge process [3]. However,

nitrification rate drops sharply as temperature of wastewater decreases and,

consequently, nitrogen removal is challenging in cool and temperate climate areas

during cold season [4].

Adsorption or ion-exchange-based approaches offer more robust alternative

method for NH4+ removal. It was recently demonstrated by a simulation that anaerobic

digestion followed by zeolite-based ion-exchange had lower operational costs and better

nitrogen-removal performance than the conventional nitrification-denitrification or the

Anammox processes [5]. Furthermore, nitrogen and carbon could be recovered in the

forms of marketable fertiliser and biogas, respectively [5]. However, it was emphasised

that the exchange capacity of the ion-exchanger material has a crucial role in the overall

process efficiency [5].

Recently, a novel NH4+ ion-exchanger material was developed: metakaolin

geopolymer [6]. The main advantages are high NH4+ exchange capacity; simple and

low-energy synthesis; and low-cost and readily available raw materials [6].

Additionally, geopolymers have been studied for the adsorption of dyes [7-10], sulphate

[11], As [12], Ca [13], Cd [14], Co [15,16], Cr [17,18], Cs [19-21], Cu [22-25], Ni

[12,26], Pb [13,17,27,28], Sb [12], Sr [19] and Zn [17]. Geopolymers consist of an

anionic framework of corner-sharing SiO4 and AlO4 where the exchangeable cations are

located in the voids – similarly as with zeolites [29]. The anion (such as arsenic or

sulphate) removal with geopolymers is based on modification or formation of secondary

mineral phases [11,12]. However, unlike zeolites, geopolymers are amorphous. The

most common synthesis method involves a reaction between aluminosilicate raw

material (such as metakaolin) and alkali activator (commonly concentrated sodium

hydroxide and silicate) at ambient or near-ambient temperature and pressure [30]. The

formation reactions of geopolymers include dissolution, gelation, reorganisation, and

hardening although the exact mechanism remains still unclear [31].

Geopolymerisation–granulation is a new method by which spherical geopolymer

granules can be produced [32]. In short, the precursor particles are mixed inside a high

shear granulator and as the alkali activator is added slowly on the particle flow, the

particles begin to bind together by the surface tension of the liquid. In addition, the

alkali activator starts to dissolve the precursor particles which enhance the binding. As

the process continues, larger and larger granules will form. The process is stopped once

the desired granule size is achieved and the granules left to harden to increase their

strength.

Zeolites, synthetic and natural, have been studied extensively for the removal

and recovery of NH4+ from various wastewaters [33-36]. Metakaolin geopolymer was

earlier noted as a more efficient NH4+ adsorbent than typical natural zeolites in terms of

capacity and kinetics [6]. However, metakaolin geopolymer has not been tested in

municipal wastewater and, in fact, almost all of the aforementioned other geopolymer

adsorbent studies have been performed with synthetic wastewaters (i.e., salts in distilled

water) as well.

Consequently, the objective of the present study was to test metakaolin

geopolymer in both synthetic and municipal wastewaters for NH4+ removal using batch

and continuous modes. Metakaolin geopolymer was prepared using a novel and easily

scalable process: geopolymerisation–granulation. Prepared granules were characterized

for their physico-chemical properties. The regenerability of the spent geopolymer was

assessed. In order to confirm the feasibility of NH4+ removal with metakaolin

geopolymer in practical conditions, a bench-scale pilot study was performed at a

wastewater treatment plant.

Materials and methods

Synthesis of geopolymers

Powdered geopolymer was synthesised by mixing metakaolin with alkali activator in a

liquid to solid weight ratio (L/S) of 1.0 for five minutes, allowed to consolidate at

ambient temperature for three days, crushed to particle size of 63–125 µm, washed with

deionised water, and stored in a desiccator before use. Alkali activator contained 12 M

sodium hydroxide (VWR Chemicals) and sodium silicate (Na2O 8.2–9.2 % w/w, SiO2 >

27.7 % w/w, VWR Chemicals) in a weight ratio of 1.2.

Granulated geopolymer was formed by mixing metakaolin powder in a high

shear granulator (Eirich EL1) and dosing the alkali activator drop-wise until an L/S ratio

of 0.4 was reached. This L/S ratio was the maximum before agglomeration of granules

started to occur. After granulation, the particle sizes of 1–4 mm (for laboratory column

experiments) or 1–6 mm (for field experiment) were separated by sieving, granules

were allowed to consolidate for three days, and material was washed with deionised

water before use.

Characterisation of geopolymers

The composition of granule cross-sections were determined by FE-SEM-EDS (field

emission scanning electron microscopy with energy dispersive X-ray spectrometer)

using Zeiss Ultra plus instrument with Oxford Instruments INCA system EDS software.

Prior the FE-SEM-EDS analysis, granules were cast in epoxy resin and polished to

reveal the cross-section. Compressive strength of granules was determined with Zwick

Roell Z010 instrument. Loose bulk density (ρb), percentage of voids (ν), apparent

particle density (ρa), oven-dried particle density (ρrd), saturated and surface-dried

particle density (ρssd) and water absorption (WA24) were determined with Equations 1–

6, respectively, according to standard methods [37,38]. = (1) = × 100 (2)

= ( ) (3)

= ( ) (4)

= ( ) (5)

= ×( ) (6)

where m2 = mass of the container and test specimen [kg], m1 = empty container [kg], V

= volume of the container [m3], ρw = density of water [kg/m3], M1 = the mass of the

saturated and surface-dried aggregate in the air [g], M2 = the apparent mass in water of

the basket containing the sample of saturated aggregate [g], M3 = the apparent mass in

water of the empty basket [g], and M4 = the mass of the oven-dried test portion in air

[g].

Laboratory-scale NH4+ removal experiments

Two wastewater samples were collected from a municipal wastewater treatment plant

(Oulu, Finland): 1) after aerated sand removal and screening (referred to as screened

effluent) and 2) after aerated sand removal, screening, coagulation with polyaluminium

chloride, and sedimentation (referred to as pre-sedimented effluent). The effects of

powdered adsorbent dose (0.5–25 g/L, 24 h contact time) and contact time (1–1440 min,

dose 5 g/L) were studied. Similar experiments were performed also with model solution

prepared of ammonium chloride (Merck). The effect of NH3–NH4+ equilibrium on the

removal was studied by adjusting the model solution pH in the range of 3–10 without

sorbent, mixing for 24 h, and comparing to the results obtained with sorbent. After each

experiment, geopolymer powder was separated by centrifuging and the supernatant was

analysed for NH4+ concentration using a flow injection analyser (Foss-Tecator Fiastar

5000).

Isotherm and kinetics models were applied to the data obtained in the batch

experiments with powdered adsorbent. Several isotherms were applied and the best-

fitting was the Sips isotherm [39] (Equation 7).

= ( )( ) (7)

where qe (mg/g) is the equilibrium sorption amount; qm (mg/g) corresponds to the

maximum sorption capacity; b (L/mg) is a parameter related to the energy of sorption;

Ce (mg/L) is the equilibrium concentration of NH4+; and the exponent n (dimensionless)

describes the heterogeneity of the sorbent surface. The kinetics data was best described

using the pseudo-second order rate equation [40] (Equation 8) which can be integrated

with the condition qt = 0 when t = 0 (Equation 9).

= ( − ) (8)

= + (9)

where qt (mg/g) is the sorption amount at time t (min), kp2 (g/(mg × min) is the pseudo-

second order rate constant.

Continuous experiments were performed by weighing 50 g of granules (1–4

mm) into a plastic column (inner height 99 mm, width 44 mm, and volume 0.15 L),

washing with deionised water, and pumping pre-sedimented effluent through the

column using flow rates of 0.5 and 1.0 L/h corresponding to 3 and 6 min empty bed

contact time (EBCT), respectively. EBCT was calculated according to Equation 10: = (10)

where Vm (L) is the volume of particles in the bed and Q (L/h) is the volumetric flow

rate.

Samples were taken with 15 min interval and NH4+ was determined as described

earlier. After filtration, the bed was flushed with 8 L of deionised water. Regeneration

of filter bed was performed by pumping 2 L of 0.1 M NaOH and 0.2 M NaCl solution

through the column using 2 L/h flow rate and rinsing with 8 L of deionised water.

Regeneration was performed two times and the NH4+ removal performance was tested

after each regeneration cycle.

Mass transfer resistance in the column experiments was assessed according to

the method described by Fulazzaky et al. [41]. In short, the internal or porous ([kLa]d),

external or film ([kLa]f), and global ([kLa]g) mass transfer coefficients (1/h) were

determined with equations 11–13, respectively. These variables were then plotted

against the percentage of outflow from the column and the curves were then used to

determine whether the mass transfer resistance is dependent on internal or external

diffusion. [ ] = [ ] − [ ] (11)

[ ] = ( ( )) (12) [ ] = [ ] × × (13)

where B (mg/g) is potential mass transfer index relating to the driving force of mass

transfer, β ((g × h)/mg) is adsorbate-adsorbent affinity parameter, C0 (mg/L) and Ct

(mg/L) are initial NH4+concentration and concentration at time t (h), and qc (mg/g) is the

cumulative adsorbed amount of NH4+ on granules. B and β were determined from the

intercept and slope, respectively, by plotting ln qc versus ln t according to Equation 14. ln = + 1/ × ln (14)

Pilot-scale NH4+ removal experiment

A small pilot-scale experiment was performed on-site at a municipal wastewater

treatment plant (Jämsä, Finland). The wastewater treatment plant process consists of

pre-treatment (screening, sand/grease separation, and pre-sedimentation), secondary

treatment (active sludge process and post-sedimentation), and tertiary treatment

(flotation and UV disinfection). Side-flow of treated wastewater (0.2 L/min) after UV

disinfection was pumped with a submersible pump through a stainless steel filter (inner

diameter 105 mm, height 600 mm and volume 5.20 L) filled with 2114g (approximately

3.2 L) of geopolymer granules (particle size 1–6 mm) (for experimental set-up, see

Figure 6). After the filter, treated water was sampled with an interval of 1 h using an

automatic sampler. The influent water to the filter was sampled manually several times

per day. NH4+ removal was monitored semiquantitatively on-site using a photometer

(Chemetrics V-2000) and cuvette tests (Chemetrics Vacu-Vials K1403) which measures

free ammonia and monochloramine [42]. The samples taken with automatic sampler

were analysed quantitatively for NH4+ with a flow injection analyser (Foss-Tecator

Fiastar 5000). Regeneration was performed by first flushing the filter bed with clean

water, pumping 25 L of 0.1 M NaOH and 0.2 M NaCl solution through the filter, and

rinsing with clean water until the effluent had low residual NH4+.

Results and discussion

Characteristics of metakaolin geopolymer

Detailed characteristics of the metakaolin geopolymer powder are presented elsewhere

[6]. In short, metakaolin geopolymer is X-ray amorphous and has higher specific

surface area and is more porous than metakaolin. Infrared spectrum, 27Si and 29Al MAS-

NMR (magic angle spinning nuclear magnetic resonance) spectra indicate a change of

chemical structure of metakaolin after geopolymerisation. Zeta potential of metakaolin

geopolymer is negative when pH > 4.5. The negative zeta potential indicates affinity

towards cations.

The cross-section of a granule (Figure 1) shows that the core (diameter of

approx. 2 mm, highlighted with white) is denser than the porous surface layer (approx.

0.5 mm). The porous nature of granules is also demonstrated by the high volume of

voids: 57.9% (Table 2). However, there are no clear differences in the chemical

composition (Table 1) across the granule indicating that geopolymerisation reaction has

taken place uniformly.

Figure 1. Cross section of metakaolin geopolymer granule. The results of point analysis

are shown in Table 1. The inner denser core has been highlighted with white colour.

Table 1. Semi-quantitative energy dispersive spectrum (EDS) analysis of points shown

in Figure 1.

Physical properties of granules are shown in Table 2. Compressive strength of granules

could not be measured in N/m2 (i.e. Pa) due to their spherical size: instead, the force (N)

needed to break granules is given. Compressive strength gives an indication of granule

quality: the obtained granules had an average strength of 63.85 N. However, there was a

large variation between individual 1–4 mm sized granules (34–123 N, n = 11). The

strength development is due to the formation of bridges between particles. It has been

noted that the plate-shapes of metakaolin particles [43] might hinder granulation and

(high) strength development [32]. Furthermore, the use of potassium hydroxide and

silicate enhances the strength development [32]. However, based on the performed

filtration experiments, the obtained strength appears to be sufficient.

Table 2. Physical properties of metakaolin geopolymer granules (1–4 mm).

NH4+ removal with powdered metakaolin geopolymer: batch experiments

The differences in the NH4+ removal efficiency from screened, pre-sedimented, and

synthetic wastewater are small: in all cases up to approx. 90% removal is reached

(Figure 2). This indicates that the geopolymer is selective towards NH4+ and wastewater

physico–chemical characteristics (Table 3) have only a minor effect on the removal

efficiency. The increase of geopolymer powder dose (Figure 2A) up to approx. 4 g/L

increases NH4+ removal results significantly but with larger doses the removal levels-off

at 85–90%. The equilibrium-state is reached after 30–90 min (Figure 2B). The rapid

adsorption has been explained to occur on the surface of the adsorbent whereas the

slower adsorption occurs inside the pores [44]. It has been shown that dosing of zeolite

powder to the activated sludge process has a beneficial effect on nitrogen removal;

zeolite acts as ion-exchager, biofilm carrier and could be regenerated continuously by

microbes; and only a small additional dosing of zeolite is necessary to replace zeolite

loss due the sludge excluding [45,46]. Since metakaolin geopolymer represents natural

zeolite chemically, a similar process could be applied with geopolymer powder. In fact,

geopolymer made of mine waste mud was found as a potential carrier media for fixed-

film wastewater-treatment processes [47].

Figure 2. NH4+ removal results from screened, pre-sedimented, and synthetic

wastewater: A) effect of geopolymer powder dose and B) effect of contact time. Initial

NH4+ was 32, 39, and 40 mg/L in synthetic, pre-sedimented and screened wastewater,

respectively.

Table 3. Physico–chemical characteristics of synthetic, pre-sedimented and screened

wastewaters.

Based on the data presented in Figure 2A, isotherm fitting was performed. Several

isotherms were applied and the best-fitting model was the Sips isotherm and thus only

its data is shown. Similarly, the best-fitting kinetics model was the pseudo-second order

rate equation. The parameters obtained from isotherm and kinetics modelling are shown

in Table 4. The maximum sorption capacity (qm) is higher than typically reported for

natural zeolites although some synthetic zeolites have still a higher capacity [6]. Lower

Si/Al ratio increases adsorption capacity as there is more negative charge in the

alumino-silicate framework and subsequently more exchangeable cations: metakaolin

geopolymer has a lower Si/Al ratio than typical natural zeolites. The maximum capacity

is only slightly reduced when comparing synthetic wastewater and screened effluent.

However, a significant decrease of qm is observed with the pre-sedimented effluent.

This could be due to the addition of flocculant and pH adjustment chemicals: for

example, calcium concentration (45 mg/L) is higher than with the screened effluent (27

mg/L) (see Table 2). Also with the kinetics, the trend of rate constants is the same:

synthetic wastewater > screened effluent > pre-sedimented effluent. Calculated and

experimental capacity values from isotherm and kinetics modelling (qm and qe,

respectively) are in agreement.

Table 4. The obtained parameters of isotherm and kinetics models.

Due to the alkaline synthesis conditions, metakaolin geopolymer increases pH even

after careful rinsing (Figure 3A). This property of geopolymers was recently utilised in

pH buffering of anaerobic digestion [48]. The increase of pH is pronounced in synthetic

wastewater due to the lack of pH buffering capacity. However, with the well-buffered

real wastewater effluents, pH increases significantly only when the dose of sorbent is

larger than 5 g/L. A blank experiment without dosing geopolymer but increasing pH

with NaOH was conducted to see the contribution of pH change and possible

volatilisation of NH3 gas on the NH4+ removal. The results of the blank experiment (not

shown) indicated that NH3 volatilisation started to contribute when pH was larger than 9

as suggested by the dissociation constant of ammonia: pKa = 9.25 at 25 °C (Figure 3B)

[49]. However, the contribution of volatilisation was rather minor (max. 6 % removal)

up to pH 11. Consequently, the volatilisation of NH3 has some effect on the removal of

NH4+ from synthetic wastewater but only a minor effect in the case of municipal

wastewater in the present experimental set-up.

Figure 3. A) change of pH as a result of sorbent dose in synthetic wastewater, screened

effluent and pre-sedimented effluent and B) speciation of NH4+ and NH3 (calculated

with MineQL+).

NH4+ removal with geopolymer granules: continuous laboratory-scale column

experiments

Metakaolin geopolymer granules were tested in column experiments using the pre-

sedimented effluent. These tests were used to preliminarily determine the effect of

empty bed contact time (EBCT): the increase of EBCT from 3 to 6 min increases the

NH4+ removal 10–20 percentage points (Figure 4A). The breakthrough point (removal

less than 50%) with EBCT of 6 min occurs at approx. 180 min (corresponding 1.56 L

outflow) and with EBCT of 3 min the removal is already ≤ 50% from the beginning.

The regeneration was performed with 0.1 M NaOH and 0.2 M NaCl solution and the

regenerated granules were used in column experiment again using 6 min EBCT (Figure

4B). The regeneration cycle was repeated twice. The regeneration was successful

although the successive regeneration cycles decreased the NH4+ removal. However, at

the beginning of the 2nd cycle (after the first regeneration) NH4+ removal was more

efficient than with virgin metakaolin geopolymer for approx. 1 h.

Figure 4. Continuous column experiments with pre-sedimented effluent: A) using empty

bed contact times (EBCT) of 6 and 3 min; B) regeneration experiments. The initial

NH4+ was 35.7 mg/L on average.

Mass transfer in granulated metakaolin geopolymer during the column test was

evaluated using the data from the series employing 6 min EBCT (Figure 4A). The plot

of ln qc versus ln t gives a straight line with a good correlation: R2 > 0.99 (Figure 5A).

Consequently, B and 1/β can be reliably obtained from the intercept and slope,

respectively (their numerical values are shown in Figure 5A). Figure 5B shows the

variation of external, internal and global mass transfer coefficients as a function of

percentage of outflow. As the granules approach saturation, the mass transfer

coefficients approach zero. The porous or internal diffusion coefficient, [kLa]d, has

negative values and thus it can be concluded that the mass transfer resistance is

dependent on internal diffusion [41]. For powdered metakaolin geopolymer, the rate-

limiting step in the NH4+ exchange process was the film diffusion as shown in our

earlier study [6].

Figure 5. A) Plot of ln qc versus ln t and B) plot of mass transfer factor (external,

internal, and global) as a function of percentage of outflow.

NH4+ removal with geopolymer granules: pilot-scale experiment

Target of the field experiment (Figure 6) was to test whether metakaolin geopolymer

granules could, in practical conditions, reduce NH4+ to a level less than 4 mg/L which is

the common discharge limit in Finland for municipal wastewater treatment plants

during the warm season (wastewater temperature >12°C). Secondly, the possibility to

regenerate metakaolin geopolymer granules was tested in practice. The influent water to

the filter had NH4+ concentration of 15–20 mg/L (Figure 7). The characteristics of

wastewater as reported by the wastewater treatment plant are shown in Table 5. During

the pilot, the NH4+ concentration was slightly lower than typically which was caused by

heavy raining. Wastewater is very suitable for treatment with metakaolin geopolymer

which is indicated by low suspended solids, BOD7,atu, CODCr, and the amount of

competing ions. Furthermore, the nitrification rate is low, i.e. most of nitrogen is present

as NH4+ in the effluent, and therefore it would be possible to reach also the 15 mg/L

total-N limit set by the EU urban wastewater directive [2].

The results of the field experiment (Figure 7) demonstrate that the discharge

limit of 4 mg/L NH4+ could be readily reached with the used set-up (flowrate 0.2 L/min

and the amount of 1–6 mm geopolymer granules 2114 g or 3.2 L). The temperature of

wastewater during field test was approx. 10 °C which indicates that the ion-exchange-

based treatment could be a feasible treatment option during the cold season. After 25 h

of run, the filter was rinsed as there was accumulation of solids in the filter although

suspended solids of influent were relatively low (Table 5). After 30 h of run, some

breakthrough of NH4+ started to occur and the 4 mg/L level was exceeded. However,

the NH4+ concentration again slowly decreased between 35–45 h of run time which

could have been due to small changes in the influent concentration. The first

regeneration was performed after 47 h of run. There was a peak of NH4+ in the effluent

after regeneration as the filter bed was rinsed. After regeneration (50–57 h of run), the

NH4+ concentration in the effluent slowly decreased. The on-site measurement of NH4

+

indicated again some breakthrough at 58 h and subsequently a second regeneration was

performed. However, the laboratory measurements revealed that the concentration was

still below 4 mg/L and the second regeneration was, in fact, unnecessary at this point.

After the second regeneration, NH4+ started to slowly decrease again: this could be due

to the leaching of residual NH4+ out of the geopolymer bed as the rinsing after

regeneration was not possibly efficient enough. The used regenerant solution had NH4+

concentration of 120 mg/L indicating that the studied process could be used to separate

and concentrate NH4+. With further concentration and selection of different regeneration

solution chemistry, the regenerant could be turned into a marketable nitrogen fertiliser

(e.g. in the form of 30% (NH4)2SO4).

Figure 6. The set-up of NH4+ removal field experiment.

Figure 7. The results of NH4+ removal field experiment.

Table 5. Physico–chemical characteristics of wastewater treated in the pilot experiment

three months before the pilot test (as reported by the wastewater treatment plant).

Conclusions

The present paper demonstrated that metakaolin geopolymer could be utilized in the

removal and possibly recovery of NH4+ from municipal wastewater. The maximum

NH4+ exchange capacity of powdered metakaolin geopolymer according to the Sips

isotherm was 31.79, 28.77, and 17.75 mg/g in synthetic, screened, and pre-sedimented

municipal wastewater, respectively. Up to approx. 90% removal of NH4+ could be

reached when the initial concentration was 32 – 40 mg/L, dose 4 g/L and contact time

60 min. The kinetics followed the pseudo-second order rate equation and rate constants

followed the trend: synthetic wastewater > screened effluent > pre-sedimented effluent.

Equilibrium was reached between 30–90 min. Geopolymerisation–granulation proved to

be a suitable and easily scalable process for producing granulated metakaolin

geopolymer. Granules had sufficient strength, porous surface, and could be regenerated

with NaCl/NaOH. However, there was some decrease in the exchange efficiency

already during two regeneration cycles. The mass transfer resistance in the continuous

column mode was determined to be due to internal diffusion. A bench-scale pilot

experiment was performed at a municipal wastewater treatment plant with tertiary

effluent in order to study the feasibility of granulated geopolymers in practical

conditions. NH4+ concentration less than 4 mg/L could be consistently reached and

regeneration was effective although wastewater temperature was only approx. 10 °C.

Metakaolin geopolymer might be interesting over zeolites due to the following reasons:

it has higher capacity than typical natural zeolites; it is less energy intensive to produce

than synthetic zeolites; and no reserves of natural zeolites are available in the Northern

Scandinavia, for example, whereas geopolymers could be prepared from locally

available materials.

Rerefences

[1] Wetzel RG. Limnology: lake and river ecosystems. 3rd ed. San Diego (CA): Elsevier; 2001.

[2] Directive 91/271/EEC of the European council concerning urban waste-water treatment. Off J Eur Union. 1991;L135;40–52.

[3] Tchobanoglous G, Burton FL, Stensel HD. Wastewater engineering: treatment and reuse. 4th ed. New York (NY): Mc Graw Hill; 2004.

[4] Jong HH, Oleszkiewicz JA. Effect of cold-temperature shock on nitrification. Water Environ Res. 2007;79:964–968.

[5] Lin Y, Guo M, Shah N, et al. Economic and environmental evaluation of nitrogen removal and recovery methods from wastewater. Bioresour Technol. 2016;215:227–238.

[6] Luukkonen T, Sarkkinen M, Kemppainen K, et al. Metakaolin geopolymer characterization and application for ammonium removal from model solutions and landfill leachate. Appl Clay Sci. 2016;119(Part 2):266–276.

[7] Li L, Wang S, Zhu Z Geopolymeric adsorbents from fly ash for dye removal from aqueous solution. J. Colloid Interface Sci. 2006;300:52–59.

[8] Khan MI, Min TK, Azizli K, et al. Effective removal of methylene blue from water using phosphoric acid based geopolymers: synthesis, characterizations and adsorption studies. RSC Adv. 2015;5:61410–61420.

[9] Yousef RI, El-Eswed B, Alshaaer M, et al. The influence of using Jordanian natural zeolite on the adsorption, physical, and mechanical properties of geopolymers products. J. Hazard. Mater. 2009;165:379–387.

[10] Liu Y, Yan C, Zhang Z, et al. A facile method for preparation of floatable and permeable fly ash-based geopolymer block. Mater Lett. 2016;185:370–373.

[11] Runtti H, Luukkonen T, Niskanen M, et al. Sulphate removal over barium-modified blast-furnace-slag geopolymer J. Hazard. Mater. 2016;317:373–384

[12] Luukkonen T, Runtti H, Niskanen M, et al. Simultaneous removal of Ni(II), As(III), and Sb(III) from spiked mine effluent with metakaolin and blast-furnace-slag geopolymers. Environ. Manage. 2016;166:579–588.

[13] Tang Q, Ge Y, Wang K, et al. Preparation and characterization of porous metakaolin-based inorganic polymer spheres as an adsorbent. Mater Des. 2015;88:1244–1249.

[14] Javadian H, Ghorbani F, Tayebi H, et al. Study of the adsorption of Cd (II) from aqueous solution using zeolite-based geopolymer, synthesized from coal fly ash; kinetic, isotherm and thermodynamic studies. Arabian J. Chem. 2015;8:837–849.

[15] Mužek MN, Svilović S, Zelić J. Kinetic studies of cobalt ion removal from aqueous solutions using fly ash-based geopolymer and zeolite NaX as sorbents. Sep. Sci. Technol. 2016;51: 2868–2875.

[16] Mužek MN, Svilovic S, Ugrina M, et al. Removal of copper and cobalt ions by fly ash-based geopolymer from solutions-equilibrium study. Desalin. Water Treat. 2016;57:10689–10699.

[17] Andrejkovicová S, Sudagar A, Rocha J, et al. The effect of natural zeolite on microstructure, mechanical and heavy metals adsorption properties of metakaolin based geopolymers. Appl. Clay. Sci. 2016;126:141–152.

[18] Cheng TW, Lee ML, Ko MS, et al. The heavy metal adsorption characteristics on metakaolin-based geopolymer. Appl. Clay. Sci. 2012;56:90–96.

[19] Chen Y, Tong Y, Pan R, et al. The research on adsorption behaviors and mechanisms of geopolymers on Sr2+, Co2+ and Cs+1,2 Advanced Mat. Res. 2013;704:313–318.

[20] López FJ, Sugita S, Kobayashi T. Cesium-adsorbent Geopolymer Foams Based on Silica from Rice Husk and Metakaolin. Chem. Lett. 2014;43:128–130.

[21] López FJ, Sugita S, Tagaya M, et al. Metakaolin-based geopolymers for targeted adsorbents to heavy metal ion separation. J. Mater. Sci. Chem. Eng. 2014;2:16–27.

[22] Ge Y, Cui X, Liao C, et al. Facile fabrication of green geopolymer/alginate hybrid spheres for efficient removal of Cu(II) in water: Batch and column studies. Chem. Eng. J. 2017;311:126–134.

[23] Alshaaer M, Zaharaki D, Komnitsas K. Microstructural characteristics and adsorption potential of a zeolitic tuff–metakaolin geopolymer. Desalin. Water Treat. 2014;56:338–345.

[24] Ge Y, Cui X, Kong Y, et al. Porous geopolymeric spheres for removal of Cu (II) from aqueous solution: Synthesis and evaluation. J. Hazard. Mater. 2015;283:244–251.

[25] Al-Harahsheh MS, Al Zboon K, Al-Makhadmeh L, et al. Fly ash based geopolymer for heavy metal removal: A case study on copper removal. J. Environ. Chem. Eng. 2015;3:1669–1677.

[26] Ge Y, Yuan Y, Wang K, et al. Preparation of geopolymer-based inorganic membrane for removing Ni2+ from wastewater. J. Hazard. Mater. 2015;299:711–718.

[27] Al-Zboon K, Al-Harahsheh MS, Hani FB. Fly ash-based geopolymer for Pb removal from aqueous solution. J. Hazard. Mater. 2011;188:414–421.

[28] Liu Y, Yan C, Zhang Z, et al. A comparative study on fly ash, geopolymer and faujasite block for Pb removal from aqueous solution. Fuel. 2016;185:181-189.

[29] Davidovits J. Geopolymer Chemistry & Applications. 3rd ed. Saint-Quentin, France: Institut Geopolymere; 2011.

[30] Luukkonen T. New Adsorption and Oxidation-Based Approaches for Water and Wastewater Treatment [dissertation]. Oulu, Finland: University of Oulu; 2016.

[31] Duxson P, Fernández-Jiménez A, Provis JL, et al. Geopolymer Technology: The Current State of the Art. J Mater Sci. 2007;42:2917–2933.

[32] Yliniemi J, Nugteren H, Illikainen M et al. Lightweight Aggregates Produced by Granulation of Peat-Wood Fly Ash with Alkali Activator. Int J Miner Process. 2016;149: 42–49.

[33] Couto RSDP, Oliveira AF, Guarino AWS, et al. Removal of ammonia nitrogen from distilled old landfill leachate by adsorption on raw and modified aluminosilicate. Environ. Technol. 2016 [cited 4.2.2017]; [11 p.]. DOI: 10.1080/09593330.2016.1212935

[34] Hedström A. Ion exchange of ammonium in zeolites: a literature review. J. Environ. Eng. 2001;127:673–681.

[35] He H, Xu S, Han R, et al. Nutrient sequestration from wastewater by using zeolite Na-P1 synthesized from coal fly ash. Environ. Technol. 2016 [cited 4.2.2017]; [8 p.]. DOI: 10.1080/09593330.2016.1217061

[36] Ye Z, Wang J, Sun L, et al. Removal of ammonium from municipal landfill leachate using natural zeolites. Environ. Technol. 2015;36:2919–2923.

[37] Finnish Standards Association (SFS). Tests for Mechanical and Physical Properties of Aggregates. Part 6: Determination of Particle Density and Water Absorption. Helsinki, Finland: SFS; 2014. Standard No. SFS-EN 1097-6.

[38] Finnish Standards Association (SFS). Tests for Mechanical and Physical Properties of Aggregates. Part 3: Determination of Loose Bulk Density and Voids. Helsinki, Finland: SFS; 1998. Standard No. SFS-EN ISO 1097-3.

[39] Sips R. On the Structure of a Catalyst Surface. J Chem Phys. 1948;16:490–495.

[40] Ho YS, McKay G. Process Biochem. Pseudo-second order model for sorption processes. 1999;34:451−465.

[41] Fulazzaky MA, Khamidun MH, Omar R. Understanding of mass transfer resistance for the adsorption of solute onto porous material from the modified mass transfer factor models. Chem. Eng. J. 2013;228:1023–1029.

[42] Krom MD. Spectrophotometric Determination of Ammonia: A Study of a Modified Berthelot Reaction using Salicylate and Dichloroisocyanurate. Analyst. 1980;105:305–316.

[43] Provis JL, Duxson P, van Deventer JSJ. The role of particle technology in developing sustainable construction materials. Adv. Powder Technol. 2010;21:2–7.

[44] Saglam N, Say R, Denizli A, et al. Biosorption of inorganic mercury and alkylmercury species on to Phanerochaete chrysosporium mycelium. Process Biochem. 1999;34:725–730.

[45] Charuckyj L, Cooksey P; Zeolite Australia Limited. Process for the removal of suspended and other material from waste water. United States patent US 9,913,012. 2004 Jan 20.

[46] Wu Z, An Y, Wang Z, et al. Study on Zeolite Enhanced Contact-Adsorption Regeneration-Stabilization Process for Nitrogen Removal. J. Hazard Mater. 2008;156:317–326.

[47] Silva I, Castro-Gomes J, Albuquerque A. Mineral waste geopolymeric artificial aggregates as alternative materials for wastewater-treatment processes: study of structural stability and pH variation in water. J. Mater. Civ. Eng. 2012;24:623–628.

[48] Novais RM, Seabra MP, Labrincha JA. Porous geopolymer spheres as novel pH buffering materials. J. Clean. Prod. 2017;143:1114–1122.

[49] Lide DR, editor. CRC Handbook of Chemistry and Physics. 84th ed. Boca Raton (FL): CRC Press; 2003.

Table 1. Semi-quantitative energy dispersive spectrum (EDS) analysis of points shown

in Figure 1.

Spectrum Label

P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11

O 50.3 48.5 45.4 48.1 45.8 52.0 48.9 46.9 46.0 42.2 48.2

Na 2.0 6.7 6.3 6.7 7.5 3.8 3.1 5.4 7.1 6.0 7.0

Mg 0.2 - 0.33 0.7 0.3 1.6 0.2 0.6 0.4 0.25 0.4

Al 7.6 18.8 17.4 17.5 18.1 10.0 22.1 11.6 18.1 14.45 18.4

Si 8.4 23.4 25.4 22.3 24.0 18.6 24.2 14.5 23.7 28.64 23.9

S - - - 0.2 - 1.2 - - - - -

K 0.5 0.8 1.6 1.9 1.7 1.7 0.7 1.1 2.5 6.39 1.0

Ca 29.7 - 0.3 0.3 0.2 5.3 - - - - -

Fe 1.2 1.7 3.2 2.2 2.4 4.6 0.7 20.0 2.2 2.1 1.2

Cu - - - - - 1.4 - - - - -

Table 2. Physical properties of metakaolin geopolymer granules (1–4 mm).

Property Value Loose bulk density, ρb 932.3 kg/m3 Percentage of voids, v 57.9 % Apparent particle density, ρa 2510 kg/m3 Oven dried particle density, ρrd 2220 kg/m3 Saturated and surface-dried particle density, ρssd 2330 kg/m3 Water absorption, WA24 5.3 % Compressive strength 63.85 N (average, n = 11)

Table 3. Physico–chemical characteristics of synthetic, pre-sedimented and screened wastewaters.

Parameter Synthetic wastewater

Pre-sedimented effluent

Screened effluent

pH 6.1 7.5 7.2

CODCr [mg/L] - 570 642

TSS [mg/L] - 419 550

NH4+ [mg/L] 32 39 40

Al [µg/L] - 8 12 P [µg/L] - 25 360 Ni [µg/L] - 6.0 5.8 Cu [µg/L] - 1.8 4.2 Zn [µg/L] - 150 170 Ca [mg/L] - 45 27 Mg [mg/L] - 5.3 4.4 Mn [mg/L] - 0.2 0.1

Table 4. The obtained parameters of isotherm and kinetics models.

Parameter Synthetic wastewater

Pre-sedimented

effluent

Screened effluent

Sips isotherm qm, experimental [mg/g] 32.00 16.59 26.40

qm, calculated [mg/g] 31.79 17.75 28.77

b [L/mg] 0.10 0.14 0.17

n 4.17 1.97 2.64

R2 0.96 0.97 0.91

RMSE 2.53 1.08 2.86

Χ2 2.29 2.45 16.18

Pseudo-second order rate equation

qe, experimental [mg/g] 5.62 5.42 5.62

qe, calculated [mg/g] 5.27 5.46 5.22

kp2 [g/(mg min)] 0.24 0.04 0.12

R2 0.97 0.99 0.98

RMSE 0.28 0.16 0.26 RMSE = residual mean square error, Χ2 = chi square test.

Table 5. Physico–chemical characteristics of wastewater treated in the pilot experiment three months before the pilot test (as reported by the wastewater treatment plant).

Parameter August 2016

September 2016

October 2016 Average

Suspended solids [mg/L] 3.4 1.4 3.2 2.7 Conductivity [mS/m] 63.6 63.0 56.6 61.1 Alkalinity [mmol/L] 1.60 0.89 0.58 1.02 pH 7.0 6.6 6.5 6.7 Total N [mg/L] 32 34 25 30

NH4+-N [mg/L] 25 22 23 23

NH4+ [mg/L] 32 28 30 30

NO3--N + NO2

--N [mg/L] 0.54 7.20 5.80 4.51

Total P [mg/L] 0.025 0.014 0.017 0.019 Soluble P [mg/L] < 0.002 0.003 < 0.002 < 0.002

BOD7,atu [mg/L] < 3 < 3 < 3 < 3

CODCr [mg/L] < 30 < 30 < 30 < 30 Al [mg/L] 0.05 0.03 0.07 0.05 Fe [mg/L] 0.091 0.100 0.057 0.083 Temperature [°C] 14.6 15.0 13.7 14.4